www.elsevier.com/locate/scitotenv

Science of the Total Environ

The Watcombe housing study: The short-term effect of improving

housing conditions on the indoor environment

George Richardson a,*, Andrew Barton b, Meryl Basham b, Chris Foy c,

Susan Ann Eick a, Margaret Somerville b

on behalf of the Torbay Healthy Housing Group, Torquay, UK

aAC and T Ltd., 12 Woolwell Drive, Plymouth, Devon, PL6 7JP, UKbPeninsula Research and Development Support Unit, N17 ITTC Building, Tamar Science Park, Plymouth, Devon, PL6 8BX, UK

cGloucestershire Research and Development Support Unit, Gloucestershire Royal Hospital, Great Western Road, Gloucester, GL1 3NN, UK

Received 13 October 2004; accepted 9 May 2005

Available online 24 June 2005

Abstract

A three-year study (1999–2001) was initiated in the UK to assess the effect of improving housing conditions in 3–4

bedroom, single-family unit, social rented sector houses on the health of the occupants. The houses were randomised into two

groups. Phase I houses received extensive upgrading including wet central heating, on demand ventilation, double-glazed doors,

cavity wall and roof/loft insulation. An identical intervention for Phase II houses was delayed for one year. As part of this

randomised waiting list study, discrete measurements were made of indoor environmental variables in each house, to assess the

short-term effects of improving housing conditions on the indoor environment. Variables representative of indoor environmental

conditions were measured in the living room, bedroom and outdoors in each of the three years of the study. In 2000, there was a

significant difference between the changes from 1999 to 2000 between Phase I (upgraded) and II (not then upgraded) houses for

bedroom temperatures ( p =0.002). Changes in wall surface dampness and wall dampness in Phase I houses were also

significantly different to the change in Phase II houses in 2000 ( p =0.001), but by 2001 the Phase I houses had reverted to

the same dampness levels they had before upgrading. The housing upgrades increased bedroom temperatures in all houses.

Other indoor environmental variables were not affected.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Indoor air quality; Social housing; Housing and health; Randomised trial; Housing upgrades

0048-9697/$ - s

doi:10.1016/j.sc

* Correspondi

E-mail addre

ment 361 (2006) 73–80

ee front matter D 2005 Elsevier B.V. All rights reserved.

itotenv.2005.05.007

ng author. Tel./fax: +44 1752 795633.

ss: ionian@tiscali.co.uk (G. Richardson).

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–8074

1. Introduction

1.1. Effect of the indoor environment on health

There is a growing understanding that the indoor

environment, particularly indoor air quality, can af-

fect health (IEH, 2001; JRC, 2003; NAS, 2000) and

that personal exposure to pollutants can often be

greater indoors than outdoors (Clayton et al.,

1993). Several indoor environmental variables are

commonly cited as having an association with health.

Cold homes in the UK have been associated with

increased cardio-respiratory mortality and morbidity

(Press, 2003). The UK Department of Health has

recommended that temperatures should be 18–21

8C in living rooms and 18 8C in bedrooms to

improve comfort and prevent health problems (DTI

and DEFRA, 2001). Dampness and relative humidity

indoors are indirectly associated with health (NAS,

2000). If relative humidity is outside the optimal

range for humans of 40–60% this can lead to health

problems linked with increases in the ideal condi-

tions for bacteria, mould and dust mites (Sterling et

al., 1985). Health problems associated with damp-

ness are mostly due to the relationship with mould

growth and other microbes. The only recommended

limit available for dampness relates to the prevention

of visible damp, mould growth or structural damage

(Protimeter plc, 2001). The limit is represented as a

wood moisture equivalent (WME%), which should

be V20%. There are no recommendations for safe

indoor levels of microbial colonies in dwellings,

despite the fact that there is evidence of negative

health effects from mould in the home (Lugauskas et

al., 2003) and known toxic properties of mould by-

products (NAS, 2000). The presence of some bacte-

rial colonies, e.g. Legionella pneumophila (Legion-

naire’s disease), indoors are a concern because of the

spread of infection.

The indoor allergen most commonly associated

with health, particularly the exacerbation of asthma

and allergies is Der p 1, from the dust mite species

Dermatophagoides pteronyssinus (NAS, 2000). There

is an international recommendation to limit Der p 1 to

V2.0 Ag g�1 to prevent sensitisation (WHO, 1988).

The World Health Organisation gives a limit for ab-

solute humidity V7.0 g kg�1 to reduce the prolifera-

tion of dust mites (WHO, 1988). Furred and feathered

pets are also a major source of allergens indoors. In

particular, cats (Felis domesticus) produce Fel d 1,

which can exacerbate asthma (van der Heide et al.,

1999).

Respirable fine particles with an aerodynamic di-

ameter b2.5 Am infiltrate from outdoors, with indoor

sources mainly originating from tobacco smoking and

cooking activities (Abt et al., 2000; BeruBe et al.,

2004). Respirable particles penetrate into the deep

lung and are associated with chronic inflammatory

processes (EPAQ, 2001). Environmental tobacco

smoke is a serious health burden, especially for chil-

dren (Kabesch and von Mutius, 2000). Inhalable

coarse particles with an aerodynamic diameter N2.5

Am, include allergenic particles, such as pollen, fungal

spores, broken up dust mite faeces and dander. No

threshold values are given in the UK for particulate

numbers, concentrations or particle size distributions

in dwellings. The UK Health and Safety Executive

has set limits for occupational exposure to dust (time

weighted averages) to 10 mg m�3 8 h for inhalable

dust and 4 mg m�3 8 h for respirable dust (HSE,

1995). Some studies have shown that there is a stron-

ger relationship between numbers of particles and

health as opposed to particle mass (Maynard, 2000),

hence the measurement of numbers of particles in this

study.

1.2. Background and description of the Watcombe

estate

The study was conducted in Watcombe, an estate

of social rented sector properties in Torquay, UK. A

self initiated survey by the residents revealed that

many of the houses had damp and mould problems

(64%) and high levels of health problems, particularly

breathing related (60%). Therefore, Torbay Council

agreed to upgrade the Watcombe houses. The Torbay

Healthy Housing Group was formed to evaluate the

impact of these non-compulsory, non-contributory

housing upgrades on the indoor environment of the

houses and the health and well being of the occupants

(Somerville et al., 2002). The estate was built during

1939–1940, in a wide shallow valley, in a coastal area

with minimal industrial airborne pollution. The houses

were of sound construction, semi-detached or ter-

raced, single-family, two storey buildings with 3–4

bedrooms (~53 m3 occupant�1).

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–80 75

The study reported here is one of the first studies

that has measured environmental and health outcomes

as part of a rigorously designed evaluation of housing

improvements in the UK.

Table 1

Environmental variables measured during assessments

Variables Units Outdoors Living

room

Bedroom

Relative humiditya % U U UTemperaturea 8C U U UFine particles

(0.3–3.0 Am)aParticle L�1

(over 30 s)

U U U

Coarse particles

(3.0–7.0 Am)aParticle L�1

(over 30 s)

U U U

Wall/wall surface

dampnessbWME % U

Dust mite allergen

Der p 1cAg g�1 U

Microbial coloniesd Number slide�1 Ua Measurements were made with a 227b Hand-held Particle Count-

er with temperature/relative humidity probe (Met One, Oregon, US);

accuracy 5% coincidence error at 70671 particle L�1 , relative

humidity F3% (10–90%) and temperature F0.4 8C (below 40 8C).b Protimeter Surveymaster, Protimeter plc, Bucks, UK; accuracy

F1%.c Collected on a filter by vacuuming 1 m2 of a mattress in the

designated bedroom for 1 min. Der p 1 extracted from the filters

using the monoclonal antibody enzyme linked immuno-sorbent

assay (ELISA).d A standard 18 cm2 dip slide was exposed to the air in the

bedroom for 1 h in each house. Dip slides were incubated at room

temperature for 240 h and the number of colonies counted.

2. Methods

2.1. Study design

Ethical approval for the study was given by the

Torbay Local Research Ethics Committee. Houses

were randomly allocated to be upgraded in 1999

(n =50, Phase I) or 2000 (n =69, Phase II). The

Phase II houses acted as a control for the Phase I

houses to assess if there had been any changes in the

indoor environment between 1999 and 2000 attribut-

able to the upgrades alone.

2.2. Intervention

The houses had an average minimum Standard

Assessment Procedure (SAP) energy rating for

dwellings (DEFRA, 2001) of 38 before upgrading.

The upgrades were aimed at improving the energy

efficiency of the houses, increasing the SAP rating

to 71–80. The upgrades consisted of wet central

heating; on demand, timed ventilation (without

heat recovery) fitted in kitchens and bathrooms;

double-glazed doors (windows already double-

glazed); cavity wall and roof/loft insulation; re-wir-

ing and re-roofing. Residents received a technical

instruction booklet explaining the correct use of the

new equipment.

2.3. Environmental data collection

A 1-h visit was made by a qualified environmental

scientist to each house in each year (January and

February; 1999–2001), during which environmental

variables from the living room, one bedroom and

outdoors were recorded using standard equipment

and discrete measurement methods described previ-

ously (Richardson et al., 2000). The bedroom was

chosen either because someone suffering from asthma

symptoms regularly slept there or because the resi-

dents identified the bedroom as damp. If neither con-

dition was met, the master bedroom was used. The

variables shown in Table 1 were chosen either because

they are associated with health problems or have an

internationally recommended dsafetyT level. The meth-

odology and measurements used were developed to

provide a representation of the indoor environment in

houses. Similar assessment protocols have been inde-

pendently developed by other researchers (Aerias,

2004; Mohle et al., 2003).

Absolute humidity (g kg�1) was calculated from

the temperature and relative humidity measurements.

Air tightness tests were not conducted but carbon

dioxide was monitored indoors and outdoors to indi-

cate ventilation levels. Air samples were collected and

analysed on a FTIR spectrometer (University of Ply-

mouth, UK) to determine carbon dioxide concentra-

tions and to check for unusual concentrations of

volatile organic compounds, formaldehyde and nitro-

gen dioxide (not reported here). Carbon monoxide

was measured in households with gas appliances—

no traces were found in any of the houses. Weather

data were obtained from the Meteorological Office

(UK).

Table 2

Comparison of Phase I and Phase II houses at the start of the study

Variable Phase I Phase II

House type: n =49 (%) n =69 (%)

Semi-detached 27 (55) 33 (48)

Terraced 14 (29) 22 (32)

Occupancy: n =48 (%) n =63 (%)

Not overcrowded

(b1 person per room)

33 (69) 49 (78)

Overcrowded

(1–1.5 people per room)

14 (29) 13 (21)

Temperature: n =48 (%) n =65 (%)

Living room b21 8C 43 (90) 58 (89)

Bedroom b18 8C 38 (79) 49 (75)

Exposure to sources of

indoor air pollutants:

n =49 (%) n =63 (%)

Pets 34 (69) 43 (68)

Smokers 37 (76) 42 (67)

NOTE: Numbers (n) vary according to response rates to individua

questions.

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–8076

2.4. Health data collection

Data on health conditions potentially influenced by

housing upgrades, or identified as problems by the

residents, were collected in an annual self-completed

postal questionnaire of each household. All adults

were interviewed by a nurse or midwife trained in

the use of the survey instruments and asked to com-

plete a SF36 questionnaire (Ware and Sherbourne,

1992) and a GHQ12 (Goldberg, 1978). Those report-

ing respiratory problems, arthritis, rheumatism, or

angina were administered condition-specific question-

naires and asked about current medication.

2.5. SAP rating

The SAP ratings were calculated for each house

before and after the upgrades, by Alba Energy Ser-

vices, Saltash, Cornwall.

2.6. Statistical analysis

The main analysis was based on a comparison

between Phase I and Phase II houses of the changes

in variables from 1999 to 2000, when only Phase I

houses had been improved. The statistical analysis was

conducted using SPSS for Windows 11.5.1. Non-para-

metric tests were used throughout, but for clarity,

means are presented in the tables. Although the pri-

mary measure was the comparison of the changes for

each phase in 2000, data is presented from the three

years of the study to clarify observed trends.

3. Results

3.1. Randomisation

Randomisation produced two groups with similar

characteristics (Table 2). Differences between the two

Phases in the following results are attributable to the

intervention, not to differences at baseline.

3.2. Response rates

Acceptance of the environmental survey was high,

with 97% (1999), 96% (2000) and 88% (2001) of

households agreeing to participate.

l

3.3. Indoor environmental outcomes: comparison be-

tween groups at 2000

The means for each variable recorded in each year

are given in Table 3. The change in bedroom tem-

perature in Phase I houses was significantly differ-

ent from that of Phase II houses between 1999 and

2000 ( p =0.002). Indoor relative humidity was

lower in 2000 than in 1999 for the Phase I houses,

but the difference between the two phases was not

significant. Wall surface and wall dampness im-

proved significantly in Phase I compared to Phase

II houses from 1999–2000 ( p =0.001), but by 2001

(approximately 15 months after upgrading) the

Phase I houses had returned to the same dampness

values they had prior to upgrading. Phase II houses

also had reduced dampness after upgrading (2000–

2001, p =0.001 for surface dampness and p =0.01

for wall dampness).

No other environmental variables changed signifi-

cantly between 1999 and 2000.

3.4. Indoor environmental outcomes: across 3 years

There was a significant reduction in the difference

between the living room and bedroom temperatures

after upgrading. The difference between living room

and bedroom temperatures reduced from 2.0 8C in

1999 to 0.7 8C in 2001 ( p b0.001). The number of

Table 4

Comparison of the mean number of fine particles indoors in each

year of the study (Phase I n=41; Phase II n=50)

Variable Phase 1999 2000 2001

(particle L�1�103)Fine particles with 1 or more I 275 263 300

smokers (living room) II 295 241 276

Fine particles with 1 or more I 244 251 275

smokers (bedroom) II 233 249 279

Fine particles with no I 181 143 248

smokers (living room) II 134 121 260

Fine particles with no I 198 127 185

smokers (bedroom) II 112 125 243

Table 3

Means for indoor environmental variables in each year of the study

Variable Phase 1999 2000 2001

Temperature living room (8C) I 19 19 19

II 18 18 19

Temperature bedroom (8C) I 16 18 18

II 16 17 19

Relative humidity living I 52 49 48

room (%) II 51 50 45

Relative humidity bedroom (%) I 56 50 51

II 56 52 46

Absolute humidity living I 7.2 6.8 6.7

room (g kg�1) II 6.7 6.9 6.2

Absolute humidity bedroom I 7.8 6.9 6.7

(g kg�1) II 7.3 6.4 6.4

Dampness (on wall surface) I 10 6 9

bedroom (WME%) II 9 9 7

Dampness (in wall fabric) I 12 6 11

bedroom (WME%) II 10 11 8

Airborne microbes bedroom I 11 10 7

(number slide�1) II 14 9 5

Coarse particles living room I 390 677 334

(particle L�1) II 331 520 402

Coarse particles bedroom I 225 567 342

(particle L�1) II 225 456 418

Fine particles living room I 255 248 290

(particle L�1�103) II 232 207 270

Fine particles bedroom I 231 228 256

(particle L�1�103) II 218 206 263

Der p 1 bedroom mattressa I - 5.22 3.73

(Ag g-1) II - 4.92 3.45

a There were no results for Der p 1 in 1999 due to technical

problems during analysis (n =97). There were no significant differ-

ences in Der p 1 concentrations between Phases I and II in 2000 or

2001.

Table 5

Means for outdoor environmental variables in each year of the study

Variable 1999 2000 2001

Precipitation (mm day�1) 2 0.1 0.1

Wind speed (m s�1) 4 8 6

Temperature (8C) 10 10 9

Relative humidity (%) 63 59 70

Absolute humidity (g kg�1) 5.1 4.7 5.3

Airborne microbial colonies (number slide�1) 9 9 6

Coarse particles (particle L�1) 84 518 177

Fine particles (particle L�1�103) 126 112 225

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–80 77

houses meeting the minimum government recommen-

dation of 18 8C rose from 23% to 75%. There were no

other significant trends.

3.5. Indoor environmental outcomes: fine particles

Means are given for Phases I and II based on

smoking status (Table 4). Fine particle concentra-

tions were substantially higher in smokers’ houses

compared to houses without smokers. The number

of fine particles in non-smokers’ houses was similar

to numbers outdoors. There were no significant

changes in the association between indoor and out-

door fine particle numbers after upgrading. Smoking

prevalence did not change in the 3 years of the

study.

3.6. Outdoor environmental outcomes

There were no significant differences in outdoor

environmental conditions between years (Table 5).

3.7. SAP ratings

The SAP ratings increased from a mean of 38

(range 28–51) to 73.5 (range 64–81) after upgrading.

3.8. Health outcomes and impact on the participants

In 2000, the changes in the health outcomes from

1999 to 2000 were not significantly different for

Phase I residents (upgraded houses) compared to resi-

dents in Phase II houses (control). The exceptions

were the prevalence of non-asthmatic respiratory ill-

ness and adult asthma symptoms, which were signif-

icantly different between phases in 2000 in favour of

the intervention. Over the 3 years of the study, asthma

prevalence reduced in children, but not in adults. The

reduction in asthma prevalence was not related to the

housing upgrades (Barton et al., unpublished data). A

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–8078

small qualitative study of tenants confirmed that more

rooms were being used following the upgrading be-

cause the temperature was more comfortable

(Basham, 2003).

4. Discussion

Over the three years of the study, the intervention

produced substantial improvements in the energy ef-

ficiency of the houses, as demonstrated by the in-

crease in SAP ratings. There was an increase in

bedroom temperatures linked to housing upgrades,

with a significant difference between phases in

2000. The housing upgrade increased the percentage

of houses meeting the minimum government recom-

mendation of 18 8C, from 23% to 75% and signifi-

cantly reduced the difference between the living room

and bedroom temperatures. The improvement in bed-

room temperatures facilitated full use of previously

dinhospitableT bedrooms. There were no clear reduc-

tions in relative humidity, mould or dampness. Out-

door relative humidity was higher in the third year,

therefore the non-significant decrease in indoor rela-

tive humidity might be linked to the intervention. The

lack of change in absolute humidity in both the living

room and bedroom suggests that if there was any

reduction in relative humidity it was due to increased

temperature, rather than a real reduction in moisture

content. The constant absolute humidity suggests that

ventilation rates did not change. Any possible reduc-

tion in involuntary ventilation does not appear to have

significantly increased the concentration of indoor

generated pollutants. The timed on-demand ventila-

tion installed was designed only to reduce localised

point sources of moisture and pollutants in the kitchen

and bathroom therefore has not had a major influence

on the air exchange in the houses. This was confirmed

by carbon dioxide measurements, which did not

change significantly after upgrading.

Dampness in the bedroom significantly decreased

in Phase I houses compared to Phase II houses after

one year, despite the fact that absolute humidity

values in the bedroom did not reduce. Dampness

mostly occurs when condensate forms on surfaces,

which are colder than the surrounding air. Therefore,

the reduction of dampness may have been due to

increased wall surface temperature created by the

cavity wall insulation rather than any changes in

humidity. In 2001, however, the dampness values in

the Phase I houses had increased to pre-upgrade

values. There was a significant increase in outdoor

relative humidity in 2001 (compared to 2000), which

coincided with the increase in dampness in Phase I

houses. However, in 2001 the Phase II houses had

reduced levels of dampness (after the upgrade in

2000), mirroring the decrease in dampness, which

occurred in Phase I houses after upgrading.

The decrease in dampness, which occurred in

Phase I houses between 1999 and 2000, does not

coincide with a reduction in outdoor humidity. In

theory, after installation of cavity wall, roof/loft insu-

lation and double-glazed doors, the internal environ-

ment of the houses should have been less influenced

by outdoor weather conditions.

It is unlikely that the differences in dampness after

upgrading were due to inaccuracies in the damp meter

or variation in measurement techniques. Although

there is an explanation as to why the upgrades

would reduce dampness (increased wall temperatures,

etc.), there is no explanation as to why dampness

would increase once again in 2001 in the Phase I

houses. This would need to be investigated with

follow-up measurements to determine if the Phase II

houses would also increase to pre-upgrade dampness

levels.

Despite incomplete data for Der p 1 concentrations

in the mattress samples, it is unlikely that there would

have been a reduction in concentrations of this asth-

ma-provoking antigen in this study, especially as there

was not a significant reduction in relative humidity in

the bedroom, which might have reduced dust mite

proliferation. Even if such a reduction had been

achieved, Der p 1 persists for a long time in the

environment, and further measures would need to

have been taken to reduce Der p 1 reservoirs in the

houses. Preventative solutions such as allergen-imper-

meable duvet, pillow and mattress covers could be

introduced. Cunningham (1998) points out that the

microclimate in bedding is very different from room

conditions. Therefore factors such as bed use, regular

airing of bedding, etc. are more important to dust mite

proliferation than general climatic conditions in the

room.

There were no significant differences between

Phase I and II houses for the number of either fine

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–80 79

or coarse particles. The overall fluctuations in the

indoor numbers of fine particles reflected changes in

outdoor numbers. It could be expected that because

the upgrades would have made the houses more dair-tightT, there would be less association between in-

door and outdoor fine particle numbers. This possi-

ble effect was not shown. Fine particle numbers are

largely determined by the amount of cigarette smoke

indoors in smokers’ houses. Smoking prevalence was

high in this population, with the majority of houses

having at least one smoker. These prevalence levels

did not change during the study, and therefore it is

not surprising that fine particle levels did not change

significantly. Coarse particles indoors are partly gen-

erated by the activities of people and pets (BeruBe et

al., 2004) and become airborne through movement.

As the thermal comfort of the houses was increased

and the tenants reported making more use of all

rooms, an increase in the number of coarse particles

in the bedroom might be expected. However, there

was no difference between the Phase I and II

changes from 1999 to 2000 in the number of coarse

particle indoors.

The number of airborne microbial colonies collect-

ed in the bedroom was largely dependent on the

ingress of microbes from outdoors and the number

of airborne spores and fragments from microbial col-

onies in the bedroom. The reduction in numbers of

microbes over the three-year period was most likely

linked to similar reductions outdoors, as there were no

differences in the changes between phases from 1999

to 2000.

The main result of this study was that there was a

more even temperature throughout the houses, with a

significant increase in bedroom temperatures. The

time between the completion of upgrades and the re-

assessment of the houses was approximately three

months, possibly this period was not long enough to

show any impact on health or other indoor environ-

ment variables. This does not detract from the fact that

by 2001 the Phase I residents had had a full year to

settle in with the upgrades. Longer follow up periods

after an intervention are suggested to determine long-

term environment and health outcomes.

From the environmental results presented here, it

could be expected that the major push to improve

housing through similar interventions may not have a

measurable impact on the indoor environment apart

from raising the indoor temperatures. Some Europe-

an countries have had energy efficient houses for

generations with no marked difference between the

health of people living in energy efficient houses

compared to energy inefficient houses. In fact, an

association is often offered of worsening health con-

ditions linked to reduced dispersion of indoor pollu-

tants in energy efficient houses (Ashmore, 1998;

Sieger et al., 1987).

The results suggest that when monitoring heating

and insulation based interventions, there is less need

for the extensive environmental assessment, per-

formed in this study. The data from this study are

being analysed with a view to identifying a reduced

environmental toolkit that can be used with less dis-

ruption and resources, but which will prove useful in

identifying changes due to heating and insulation type

interventions.

Although information was provided for the resi-

dents on the use of their new equipment, the informa-

tion was based on a technical guide rather than

addressing the needs of the residents. Specifically

tailored information is needed for people to help

them understand the nature of a healthy indoor envi-

ronment and how they can achieve that most efficient-

ly in their own home.

5. Conclusions

The housing upgrades produced a substantial

increase in the energy efficiency of the houses

and an improvement in thermal comfort as an im-

mediate result. The extent to which such upgrades

can be expected to improve the indoor environment

may be limited, as occupants, their habits and in-

door activities remain substantially the same and

influence the variables measured. Improving hous-

ing through energy efficiency interventions may not

have a measurable impact on the indoor environ-

ment apart from raising indoor temperatures. Nev-

ertheless, it remains important to improve the

thermal efficiency of homes to save energy, thereby

reducing CO2 emissions, reduce annual fuel costs

for householders and to improve comfort and the

standard of living within a home. This study

demonstrates that more tailored interventions are

needed to impact on the indoor environment to

G. Richardson et al. / Science of the Total Environment 361 (2006) 73–8080

directly influence health. Longer follow-up periods

may be needed in order to understand the effect of

housing upgrades on the interaction between people

and their indoor environment.

Acknowledgements

The authors would like to thank the residents of the

Watcombe estate for their time and patience in partic-

ipating in this study and the NHS (SW) R and D for

funding the evaluation of the housing upgrades.

References

Abt E, Suh HH, Allen G, Koutrakis P. Characterization of indoor

particle sources: a study conducted in the Metropolitan Bos-

ton area. Environ Health Perspect 2000;108(1):35–44.

Aerias. Indoor air quality investigations in the workplace and

other buildings. Aerias: air quality sciences. IAQ Resource

Centre. Available: http://www.aerias.org/DesktopDefault.aspx?

tabindex=3 and tabid=79 (Accessed 8 September 2004).

Ashmore I. Asthma, housing and environmental health. Environ

Health 1998;106(1):17–24.

Basham M. Qualitative study of central heating—its influence on

the use of the house, the behaviour and relationships of the

household particularly in wintertime. Proceedings of the con-

ference on (Un)healthy housing: promoting good health, 19–21

March. Coventry, UK7 University of Warwick; 2003.

BeruBe KA, Sexton KJ, Jones TP, Moreno T, Anderson S, Richards

RJ. The spatial and temporal variations in PM10 mass from six

homes. Sci Total Environ 2004;324:41–53.

Clayton CA, Peritt RL, Pellizzari ED, Thomas KW, Whitmore RW,

Wallace LA, et al. Particle total exposure assessment method-

ology (PTEAM) study: distributions of aerosol and elemental

concentrations in personal, indoor and outdoor air samples in a

Southern Californian Community. J Expo Anal Environ Epide-

miol 1993;3:227–50.

Cunningham MJ. Direct measurements of temperature and humid-

ity in dust mite microhabitats. Clin Exp Allergy 1998;28:

1104–12.

DEFRA. Government standard assessment procedure for energy

rating of dwellings, Department for Environment, Food and

Rural Affairs. London, UK7 HMSO; 2001.

DTI, DEFRA. The UK fuel poverty strategy, Department of Trade

and Industry and Department for Environment, Food and Rural

Affairs. London, UK7 HMSO; 2001.

EPAQ. Airborne particles: what is the appropriate measurement on

which to base a standard? Expert panel on air quality standards.

London, UK7 DEFRA; 2001.

Goldberg D. General health questionnaire, GHQ12. Berkshire, UK7

The NFER-NELSON Publishing Company Ltd; 1978.

HSE. EH44 dust: general principles of protection. Health and Safety

Executive. ISBN 0717614352.

IEH. Indoor air quality in the home: final report on DETR contract

EPG 1/5/12 (web report W7). Institute for Environment and

Health. Available: http://www.le.ac.uk/ieh/pdf/w7.pdf (Accessed

11 October 2004).

JRC. Indoor air pollution: new EU research reveals higher risks than

previously thought. European Commission Joint Research Cen-

tre. Available: http://www.jrc.cec.eu.int/ (Accessed 11 October

2004).

Kabesch M, von Mutius E. Adverse health effects of environmental

tobacco smoke exposure in childhood. Allergy Clin Immunol

Int 2000;12(4):146–52.

Lugauskas A, Krikstaponis A, Seskauskas V. Species of con-

ditionally pathogenic micromycetes in the air of dwellings

and occupational premises. Indoor Built Environ 2003;12(3):

167–77.

Maynard RL. New directions: reducing the toxicity of vehicle

exhaust. Atmos Environ 2000;34:2667–8.

Mohle G, Crump D, Brown V, Hunter C, Squire R, Mann H, et al.

Development and application of a protocol for the assessment of

indoor air quality. Indoor Built Environ 2003;12:139–49.

NAS. Clearing the air: asthma and indoor air exposures. National

Academy of Sciences, Institute of Medicine. Washington DC,

US: National Academic Press; 2000.

Press V. Fuel poverty+health: a guide for primary care organisations

and public health and primary care professionals. London, UK7

National Heart Forum; 2003.

Protimeter plc. Protimeter moisture measurement solutions. Proti-

meter plc, UK; 2001.

Richardson G, Eick SA, Harwood DJ. Quantifying ventilation

needs in local authority housing—a case study. Proceedings

of the 21st AIVC Conference, Innovations in Ventilation Tech-

nology, The Hague, Netherlands, 26–29 September; 2000.

paper 46.

Sieger TL, Fiore MC, Anderson HA, Ziarnik ME, Bush RK, Dopico

GA, et al. The health effects and environmental assessment of

btightQ homes. Proceedings of the 80th Annual meeting of

APCA, New York, USA, 21–26 June; 1987.

Somerville M, Basham M, Foy C, Ballinger G, Gay T, Shute P, et al.

From local concern to randomised trial: the Watcombe housing

project. Health Expect 2002;5(2):127–35.

Sterling EM, Arundel A, Sterling TD. Criteria for human exposure

to humidity in occupational buildings. ASHRAE Trans 1985;

91:611–20.

van der Heide S, van Aalderen WM, Kaufman HF, Dubois AE, de

Monchy JG. Clinical effects of air cleaners in homes of asth-

matic children sensitised to pet allergens. J Allergy Clinl Immu-

nol 1999;104(2Pt 1):447–51.

Ware JE, Sherbourne CD. The MOS 36-item short-form health

survey (SF-36): conceptual framework and item selection.

Med Care 1992;30:473–83.

WHO. Dust mite allergen and asthma: a world-wide problem. Bull

W H O 1988;66(6):769–80.